Ignorance and Decision-Making, Part Five
Why think that epistemic considerations must be logically, rather than merely psychologically, prior to other considerations in decision-making?
This is the fifth and, for the time being, final part of my multi-part series on ignorance and decision-making. I will probably update the series from time to time as the project continues to develop. As I mentioned last week, we have another paper on our second round of experiments currently under review and the data from a third round of experiments still to analyze. So, I will probably post additional newsletters in the future to address these and other developments. If I ever finish the series once and for all, I will try to collect all of the posts into a single newsletter for anyone who might want to read all of it in one sitting.
To say that some set of factors is logically prior in decision-making to some other set of factors is to say that the latter cannot play a causal role in decision-making unless the former plays its causal role.
Logical priority is a (logically) stronger kind of priority than other kinds of priority that might figure in decision-making, e.g., psychological priority. If some set of factors is logically prior, it is also psychologically prior, but not necessarily vice versa. What’s more, causal factors that are merely psychologically prior are only contingently so. The evolutionary considerations from which the psychological priority of some factors emerged could have been different than they were and, if they had been different, some other set of factors might have emerged as psychologically basic. But, if some set of factors is logically prior to some other, they are necessarily so. They could not have failed to be psychologically basic, whatever the evolutionary considerations to which they were subjected.
There are good reasons to think that epistemic burdens – the nature and extent of a decision-maker’s relevant ignorance in some decision context – are logically prior in decision-making and not merely psychologically prior.
Imagine that, rather than epistemic criteria serving to cull and sort courses of action, the evolutionary considerations to which decision-making psychology was subjected were such that courses of action came to be sorted into a cognitively tractable preference ranking according to the nature and extent of the sacrifice required to successfully pursue them. More carefully, imagine that courses of action carrying lower opportunity costs were always ranked higher in a decision-maker’s preference ranking, other things equal, than other courses of action carrying higher opportunity costs (while courses of action carrying infinitely high opportunity costs were simply culled from, not included in, the decision-makers’ preference ranking). If this were the case, then relevant opportunity costs would be psychologically prior to other causal factors, e.g., the ethical, prudential, or pecuniary values of particular courses of action, that could have played the same function in decision-making had evolutionary processes and relevant considerations been different.
Yet, even if this had been the case, decision-makers would still struggle to make decisions unless they (consciously or otherwise) knew, could evaluate and compare, the relevant opportunity costs associated with different courses of action. In other words, even if evolution had been such that some non-epistemic factor had come to serve as criteria for the culling and sorting of courses of action into a tractable menu of options, relevant knowledge would nevertheless have to come first in decision-making, for these non-epistemic criteria to play this role.
Epistemic considerations must be logically prior, not merely psychologically prior, to non-epistemic considerations in decision-making.
All of this granted, however, experimental evidence of the sort discussed in the fourth part of this series, would seem to be powerless, in isolation, to differentiate logical priority from mere psychological priority. No experiment can conclusively indicate whether epistemic considerations are necessarily rather than merely contingently prior in decision-making.
However, the foregoing thought experiment, in conjunction with the philosophical arguments discussed in the second part of the series, the illustration of the role of epistemic burdens in surrogate decision-making described in the third part of the series, and the aforementioned experimental evidence, conspire to support the truth of the stronger thesis.
Throughout this series, I have carefully avoided the implication that the logical priority of the epistemic is meant to apply only to human decision-makers. I should probably explicate at this point that the logical priority of the epistemic is posited to apply to all decision-makers, human and otherwise. For a dog to bury a bone, the dog must (in some sense specific to canine epistemology) know there is something to be buried. For a cat to kill a mouse, it must (in some sense…) know there is something to be killed.
If this is right, if the priority of the epistemic applies to all decision-makers, human and otherwise, it is yet another reason to think that the relevant kind of priority is logical and not merely psychological. Otherwise, it would be an unexplained miracle that different species with unique biological histories, unique anatomies and physiologies, that have been subject to different evolutionary circumstances over time, all evolved in such a way that epistemic burdens became merely psychologically, but not logically prior, in their unique, species-specific, decision-making processes.